def testTensorShapeUnknown(self):
context = function_trace_type.SignatureContext()
spec_1 = tensor_spec.TensorSpec(
None, dtype=dtypes.int32)._tf_tracing_type(context)
spec_2 = tensor_spec.TensorSpec(
None, dtype=dtypes.int32)._tf_tracing_type(context)
self.assertEqual(spec_1, spec_2)
def testTensorAndSpecEquality(self):
context = function_trace_type.SignatureContext()
tensor = array_ops.zeros([11, 3, 5],
dtype=dtypes.int32)._tf_tracing_type(context)
spec = tensor_spec.TensorSpec([11, 3, 5],
dtype=dtypes.int32)._tf_tracing_type(context)
spec_with_name = tensor_spec.TensorSpec(
[11, 3, 5], dtype=dtypes.int32, name='name')._tf_tracing_type(context)
self.assertEqual(tensor, spec)
self.assertNotEqual(tensor, spec_with_name)
def make_cache_key(
args,
include_tensor_ranks_only: bool = False
) -> Tuple[FunctionCacheKey, function_trace_type.WeakrefDeletionObserver]:
"""Computes the cache key given the function arguments."""
signature_context = function_trace_type.SignatureContext(
include_tensor_ranks_only)
function_signature = function_trace_type.make_function_signature(
args, signature_context)
return FunctionCacheKey(
function_signature,
_make_execution_context()), signature_context.deletion_observer
def testTensorEquality(self):
context = function_trace_type.SignatureContext()
tensor_a = array_ops.zeros(
[11, 3, 5], dtype=dtypes.int32)._tf_tracing_type(context)
tensor_b = array_ops.zeros(
[11, 4, 5], dtype=dtypes.int32)._tf_tracing_type(context)
tensor_c = array_ops.zeros(
[11, 3, 5], dtype=dtypes.float32)._tf_tracing_type(context)
tensor_d = array_ops.ones([11, 3, 5],
dtype=dtypes.int32)._tf_tracing_type(context)
self.assertNotEqual(tensor_a, tensor_b)
self.assertNotEqual(tensor_a, tensor_c)
self.assertNotEqual(tensor_b, tensor_c)
self.assertEqual(tensor_a, tensor_d)
def make_function_signature_with_context(inputs):
return function_trace_type.make_function_signature(
inputs, function_trace_type.SignatureContext())
